Checking Distribution for Variables

Plot and fit distributions of variables

Histogram of Velocity of all

Log distribution, 0’s make up 3% of all data

sd(log10(CC.TotalData$v[CC.TotalData$v>0]))

[1] 0.543003

Now we’re going to plot each unique krill and their bimodality with a dip test (less than 0.05 is multimodality)

ind <- unique(CC.TotalData$D_V_T)
print(ind)

  [1] 20191118_1__1 20191119_1__4 20191119_1__5 20191119_1__1 20191119_1__2 20191119_1__3 20191119_2__1 20191119_2__2 20191119_2__4 20191119_2__3
 [11] 20191119_2__5 20191119_3__1 20191119_3__2 20191119_3__5 20191119_3__4 20191119_3__3 20191119_5__1 20191119_5__3 20191119_5__4 20191119_5__2
 [21] 20191119_6__1 20191119_6__3 20191119_7__1 20191119_7__2 20191119_7__3 20191119_7__5 20191119_7__4 20191120_2__1 20191120_2__2 20191120_2__3
 [31] 20191120_2__4 20191120_2__5 20191120_4__2 20191120_4__4 20191120_4__3 20191120_4__1 20191121_1__2 20191121_1__3 20191121_1__4 20191121_1__1
 [41] 20191121_2__3 20191121_2__4 20191121_2__5 20191121_2__6 20191121_2__1 20191121_2__2 20191121_4__1 20191121_4__2 20191121_4__3 20191121_4__4
 [51] 20191121_5__3 20191121_5__4 20191121_5__2 20191121_5__1 20191121_6__1 20191121_6__3 20191121_6__2 20191121_6__4 20191121_7__1 20191121_7__2
 [61] 20191121_7__3 20191121_7__4 20191122_2__2 20191122_2__1 20191122_2__3 20191122_2__4 20191122_3__2 20191122_3__3 20191122_3__4 20191122_3__1
 [71] 20191124_10_1 20191124_10_2 20191127_1__3 20191127_1__4 20191127_1__6 20191127_1__7 20191127_1__5 20191127_1__1 20191127_1__2 20191127_4__2
 [81] 20191127_4__1 20191127_4__3 20191127_4__4 20191127_4__5 20191127_4__6 20191129_2__1 20191129_2__2 20191129_2__3 20191129_2__4 20191129_3__1
 [91] 20191129_3__2 20191129_3__3 20191129_3__4 20191129_6__1 20191129_6__2 20191129_6__3 20191129_6__4 20191130_1__1 20191130_1__2 20191130_1__3
[101] 20191130_1__4 20191130_3__1 20191130_3__2 20191130_3__3 20191130_3__4 20191130_6__1 20191130_6__2 20191130_6__3 20191130_6__4 20191130_6__5
[111] 20191201_2__1 20191201_2__2 20191201_2__3 20191201_2__4 20191202_1__1 20191202_1__2 20191202_1__3 20191202_1__4 20191204_2__1 20191204_2__2
[121] 20191204_2__3 20191204_2__4 20191205_10_1 20191205_10_2 20191205_10_3 20191205_10_4 20191212_3__1 20191212_3__2 20191212_3__3
153 Levels: 20191118_1__1 20191119_1__1 20191119_1__2 20191119_1__3 20191119_1__4 20191119_1__5 20191119_2__1 20191119_2__2 20191119_2__3 ... 20191212_3__4

length(ind)

[1] 129

library(diptest)
library(DescTools)

tab <- matrix(data = NA, nrow = 129, ncol =5, byrow = T)
colnames(tab) <- c('dip.test', 'skew', 'mean.velocity', 'sd.velocity', 'Ind')

tab <- as.data.frame(tab)
tab$Ind <- ind
mean.vel <- NULL
sd.vel <- NULL
d.v <- NULL
s.v <- NULL

for (i in 1:length(ind)){
mean.v <- mean(CC.TotalData$v[CC.TotalData$D_V_T==ind[i]])
sd.v <- sd(CC.TotalData$v[CC.TotalData$D_V_T==ind[i]])

mean.vel <- rbind(mean.vel, mean.v)
sd.vel <- rbind(sd.vel, sd.v)


vels <- (CC.TotalData$v[CC.TotalData$D_V_T==ind[i]])
vels <- log10(vels[vels>0])
d <- dip.test(vels)
d.p <- d$p.value
d.v <- rbind(d.v, d.p)

s <- Skew(vels)
s.v <- rbind(s.v, s)
##}

##ind <- unique(CC.TotalData$D_V_T)

##for (i in 1:length(ind)){
 hist(log10(CC.TotalData$v[CC.TotalData$D_V_T==ind[i]]),
     breaks = 50,
     xlab = "Velocity (Log^10 mm/s)",
     main = ind[i],
     sub = d.p)  ## change to d.p or s to print the dip test or skew value as the title instead
}

tab$skew <- s.v
tab$dip.test <- d.v
tab$mean.velocity <- (log10(mean.vel))
tab$sd.velocity <- (log10(sd.vel))
tab

write.table(tab, file = "~/Post-doc/Data/dip.test.skew.vels.csv", sep = ",", col.names = TRUE)

plot(tab)

head(TotalData)

str(TotalData)

'data.frame':   1722390 obs. of  23 variables:
 $ Date       : chr  "20191118" "20191118" "20191118" "20191118" ...
 $ File.name  : chr  "20191118_view1_" "20191118_view1_" "20191118_view1_" "20191118_view1_" ...
 $ X          : num  0.319 NaN 0.318 NaN 0.343 ...
 $ Y          : num  NaN NaN NaN NaN 0.0999 ...
 $ Z          : num  NaN -0.0573 NaN -0.0567 -0.055 ...
 $ Track      : int  1 1 1 1 1 1 1 1 1 1 ...
 $ View       : chr  "1_" "1_" "1_" "1_" ...
 $ D_V_T      : Factor w/ 153 levels "20191118_1__1",..: 1 1 1 1 1 1 1 1 1 1 ...
 $ D_V        : Factor w/ 35 levels "20191118_1_",..: 1 1 1 1 1 1 1 1 1 1 ...
 $ Flow.rate  : num  0 0 0 0 0 0 0 0 0 0 ...
 $ Chlorophyll: num  0 0 0 0 0 0 0 0 0 0 ...
 $ Guano      : Factor w/ 2 levels "Absent","Present": 1 1 1 1 1 1 1 1 1 1 ...
 $ Light      : Factor w/ 2 levels "Absent","Present": 2 2 2 2 2 2 2 2 2 2 ...
 $ dx         : num  NaN NaN NaN NaN -0.00115 ...
 $ dy         : num  NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN ...
 $ dz         : num  NaN NaN NaN 0.00168 NaN ...
 $ d          : num  NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN ...
 $ vx         : num  NaN NaN NaN NaN -0.0345 ...
 $ vy         : num  NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN ...
 $ vz         : num  NaN NaN NaN 0.0503 NaN ...
 $ v          : num  NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN ...
 $ heading    : num  NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN ...
 $ pitch      : num  NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN ...

TotalData$Flow.rate <- as.factor(TotalData$Flow.rate)
TotalData$Chlorophyll <- as.factor(TotalData$Chlorophyll)

library(ggplot2)

package 㤼㸱ggplot2㤼㸲 was built under R version 3.6.3

ggplot(TotalData,aes(x=Flow.rate, y=turn.angle, fill=Chlorophyll))+
  geom_boxplot(notch=F, notchwidth=0.3,outlier.shape=1,outlier.size=2, coef=1.5)+
  theme(axis.text=element_text(color="black"))+
  theme(axis.text.x=element_text(angle=90,hjust=1,vjust=0.4))+
  theme(panel.grid.minor=element_blank())+
  labs(size= "",x = "Flow Rate (cm/s)", y = "Turn angle (degrees)", title = "               Light")+
  scale_fill_manual(values=c("greenyellow", "green","green1", "green2", "green3", "green4"),name = "Chlorophyll (mg/L)",
                    labels=c("0", "4.6", "6.1", "7.6", "13.5", "19"))+
  facet_grid(~Light, scales = "free_x", space = "free")

Error in FUN(X[[i]], ...) : object 'turn.angle' not found

Now looking at turning angles

Starting to look at bimodality in the variables by merging the TotalData frame with the “tab” table

print (prob)

     
               0        4.3        4.6        6.1        7.6       13.5         19
  0   0.37209302 0.00000000 0.00000000 0.00000000 0.09302326 0.00000000 0.10077519
  0.6 0.01550388 0.03100775 0.03100775 0.03100775 0.00000000 0.02325581 0.00000000
  3   0.03100775 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000
  5.9 0.17829457 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000
  8.9 0.09302326 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000

save.image("~/Post-doc/Data/Total Merged Data File.RData")